Project 5. Project Summary. Project 5 will conduct behavioral and fMRI studies in people who have focal lesions to the prefrontal cortex, recruited from the lesion registry at the University of Iowa. It is a major extension of our current Conte Center, and has a subcontract to PI Daniel Tranel at Iowa. Its three Aims focus on dissociation of basic decision-making systems (continuous with Project 1 in our current Conte Center), studies of social inference (the same tasks as under the renewal Project 2, Aim 1), and fMRI studies probing compensatory processing. Given the strong links to other Projects, it also includes as personnel several PIs from other Projects (Adolphs, O'Doherty) as well as shared post-docs. The overarching goal of Project 5 is to use the lesion method to investigate the necessary role of specific sectors of the prefrontal cortex, the brain region most important for social decision-making. We are particularly interested in medial regions of the PFC, known to be critical for representing value and social inference, and in dorsolateral prefrontal cortex, known to be important for context-dependent regulation. How damage to these regions may dissociate specific components of the social decision-making process, let alone how it may engage compensatory processing through remaining intact network components, is largely unexplored. This will be a crucially important complement to all other studies under this Conte Center, since it provides a window into the necessary role of brain regions. Aim 1 focuses on the role of the ventromedial prefrontal cortex in goal-directed or habit-based control in instrumental choice. We will build on strong pilot data that shows, for the first time in humans, that the vmPFC is essential for goal-directed, but not habit-based choice. Aim 2 will use the full battery of tasks developed under Aim 1 of Project 2 to probe the necessary role of the prefrontal cortex in social inference. This task battery probes inferences made to social or nonsocial stimuli, as a function of cognitive load, to facial expressions or hand actions, and other factors. Finally, Aim 3 will analyze the tasks under Aims 1 and 2 done in fMRI experiments, to examine the neural regions responsible for possible compensatory task performances. The studies will be executed in 30 patients with focal, chronic lesions to the prefrontal cortex, 30 subjects with lesions elsewhere, and 30 healthy comparison subjects, using a combination of ROI-based and voxel-based lesion mapping. Some of the ROIs we will specifically query are those generated from the fMRI studies in Projects 1, 2 and 3: those fMRI studies tell us which regions are activated in healthy brains, but only Project 5 can tell us if damage within those same regions also compromises social inference and decision- making performances.